| Message ID | 1450169379-12336-8-git-send-email-zhaoshenglong@huawei.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
On Tue, Dec 15, 2015 at 04:49:27PM +0800, Shannon Zhao wrote: > From: Shannon Zhao <shannon.zhao@linaro.org> > > When we use tools like perf on host, perf passes the event type and the > id of this event type category to kernel, then kernel will map them to > hardware event number and write this number to PMU PMEVTYPER<n>_EL0 > register. When getting the event number in KVM, directly use raw event > type to create a perf_event for it. > > Signed-off-by: Shannon Zhao <shannon.zhao@linaro.org> > --- > arch/arm64/include/asm/pmu.h | 3 ++ > arch/arm64/kvm/Makefile | 1 + > include/kvm/arm_pmu.h | 11 ++++ > virt/kvm/arm/pmu.c | 122 +++++++++++++++++++++++++++++++++++++++++++ > 4 files changed, 137 insertions(+) > create mode 100644 virt/kvm/arm/pmu.c [...] > +/** > + * kvm_pmu_set_counter_event_type - set selected counter to monitor some event > + * @vcpu: The vcpu pointer > + * @data: The data guest writes to PMXEVTYPER_EL0 > + * @select_idx: The number of selected counter > + * > + * When OS accesses PMXEVTYPER_EL0, that means it wants to set a PMC to count an > + * event with given hardware event number. Here we call perf_event API to > + * emulate this action and create a kernel perf event for it. > + */ > +void kvm_pmu_set_counter_event_type(struct kvm_vcpu *vcpu, u64 data, > + u64 select_idx) > +{ > + struct kvm_pmu *pmu = &vcpu->arch.pmu; > + struct kvm_pmc *pmc = &pmu->pmc[select_idx]; > + struct perf_event *event; > + struct perf_event_attr attr; > + u64 eventsel, counter; > + > + kvm_pmu_stop_counter(vcpu, pmc); > + eventsel = data & ARMV8_EVTYPE_EVENT; > + > + memset(&attr, 0, sizeof(struct perf_event_attr)); > + attr.type = PERF_TYPE_RAW; > + attr.size = sizeof(attr); > + attr.pinned = 1; > + attr.disabled = kvm_pmu_counter_is_enabled(vcpu, select_idx); > + attr.exclude_user = data & ARMV8_EXCLUDE_EL0 ? 1 : 0; > + attr.exclude_kernel = data & ARMV8_EXCLUDE_EL1 ? 1 : 0; > + attr.exclude_hv = 1; /* Don't count EL2 events */ > + attr.exclude_host = 1; /* Don't count host events */ > + attr.config = eventsel; > + > + counter = kvm_pmu_get_counter_value(vcpu, select_idx); > + /* The initial sample period (overflow count) of an event. */ > + attr.sample_period = (-counter) & pmc->bitmask; > + > + event = perf_event_create_kernel_counter(&attr, -1, current, NULL, pmc); As far as I can see, this is going to result in unreliable counts on a big.LITTLE system, even if the VCPUs are constrained to one class of core. As this is a task-bound event (cpu == -1, task is current), the perf core will stop as soon as one PMU driver agrees to handle the event. The event will then only count on CPUs handled by that driver. If you're unlucky, the set of CPUs handled by that driver is not the same as the set of CPUs your VM is constrained to. e.g. your VM might be on little cores, but the big PMU driver accepted the event, and only counts on big cores. I'm not sure how we can solve that. Thanks, Mark. > + if (IS_ERR(event)) { > + printk_once("kvm: pmu event creation failed %ld\n", > + PTR_ERR(event)); > + return; > + } > + > + pmc->perf_event = event; > +} > -- > 2.0.4 > > > _______________________________________________ > kvmarm mailing list > kvmarm@lists.cs.columbia.edu > https://lists.cs.columbia.edu/mailman/listinfo/kvmarm >
On 17/12/15 15:22, Mark Rutland wrote: > On Tue, Dec 15, 2015 at 04:49:27PM +0800, Shannon Zhao wrote: >> From: Shannon Zhao <shannon.zhao@linaro.org> >> >> When we use tools like perf on host, perf passes the event type and the >> id of this event type category to kernel, then kernel will map them to >> hardware event number and write this number to PMU PMEVTYPER<n>_EL0 >> register. When getting the event number in KVM, directly use raw event >> type to create a perf_event for it. >> >> Signed-off-by: Shannon Zhao <shannon.zhao@linaro.org> >> --- >> arch/arm64/include/asm/pmu.h | 3 ++ >> arch/arm64/kvm/Makefile | 1 + >> include/kvm/arm_pmu.h | 11 ++++ >> virt/kvm/arm/pmu.c | 122 +++++++++++++++++++++++++++++++++++++++++++ >> 4 files changed, 137 insertions(+) >> create mode 100644 virt/kvm/arm/pmu.c > > [...] > >> +/** >> + * kvm_pmu_set_counter_event_type - set selected counter to monitor some event >> + * @vcpu: The vcpu pointer >> + * @data: The data guest writes to PMXEVTYPER_EL0 >> + * @select_idx: The number of selected counter >> + * >> + * When OS accesses PMXEVTYPER_EL0, that means it wants to set a PMC to count an >> + * event with given hardware event number. Here we call perf_event API to >> + * emulate this action and create a kernel perf event for it. >> + */ >> +void kvm_pmu_set_counter_event_type(struct kvm_vcpu *vcpu, u64 data, >> + u64 select_idx) >> +{ >> + struct kvm_pmu *pmu = &vcpu->arch.pmu; >> + struct kvm_pmc *pmc = &pmu->pmc[select_idx]; >> + struct perf_event *event; >> + struct perf_event_attr attr; >> + u64 eventsel, counter; >> + >> + kvm_pmu_stop_counter(vcpu, pmc); >> + eventsel = data & ARMV8_EVTYPE_EVENT; >> + >> + memset(&attr, 0, sizeof(struct perf_event_attr)); >> + attr.type = PERF_TYPE_RAW; >> + attr.size = sizeof(attr); >> + attr.pinned = 1; >> + attr.disabled = kvm_pmu_counter_is_enabled(vcpu, select_idx); >> + attr.exclude_user = data & ARMV8_EXCLUDE_EL0 ? 1 : 0; >> + attr.exclude_kernel = data & ARMV8_EXCLUDE_EL1 ? 1 : 0; >> + attr.exclude_hv = 1; /* Don't count EL2 events */ >> + attr.exclude_host = 1; /* Don't count host events */ >> + attr.config = eventsel; >> + >> + counter = kvm_pmu_get_counter_value(vcpu, select_idx); >> + /* The initial sample period (overflow count) of an event. */ >> + attr.sample_period = (-counter) & pmc->bitmask; >> + >> + event = perf_event_create_kernel_counter(&attr, -1, current, NULL, pmc); > > As far as I can see, this is going to result in unreliable counts on a > big.LITTLE system, even if the VCPUs are constrained to one class of > core. > > As this is a task-bound event (cpu == -1, task is current), the perf > core will stop as soon as one PMU driver agrees to handle the event. The > event will then only count on CPUs handled by that driver. > > If you're unlucky, the set of CPUs handled by that driver is not the > same as the set of CPUs your VM is constrained to. e.g. your VM might be > on little cores, but the big PMU driver accepted the event, and only > counts on big cores. > > I'm not sure how we can solve that. Yeah, another level of BL braindeadness. We should have some API to match the PMU we want on the CPU we're on at the moment that trap occurs. I don't think this should block this series though - this is something we can improve on in parallel (a possible solution being to forbid KVM on BL platform altogether). Thanks, M.
diff --git a/arch/arm64/include/asm/pmu.h b/arch/arm64/include/asm/pmu.h index 4264ea0..714edcc 100644 --- a/arch/arm64/include/asm/pmu.h +++ b/arch/arm64/include/asm/pmu.h @@ -18,6 +18,7 @@ #define ARMV8_MAX_COUNTERS 32 #define ARMV8_COUNTER_MASK (ARMV8_MAX_COUNTERS - 1) +#define ARMV8_CYCLE_IDX (ARMV8_MAX_COUNTERS - 1) /* * Per-CPU PMCR: config reg @@ -28,6 +29,8 @@ #define ARMV8_PMCR_D (1 << 3) /* CCNT counts every 64th cpu cycle */ #define ARMV8_PMCR_X (1 << 4) /* Export to ETM */ #define ARMV8_PMCR_DP (1 << 5) /* Disable CCNT if non-invasive debug*/ +/* Determines which PMCCNTR_EL0 bit generates an overflow */ +#define ARMV8_PMCR_LC (1 << 6) #define ARMV8_PMCR_N_SHIFT 11 /* Number of counters supported */ #define ARMV8_PMCR_N_MASK 0x1f #define ARMV8_PMCR_MASK 0x3f /* Mask for writable bits */ diff --git a/arch/arm64/kvm/Makefile b/arch/arm64/kvm/Makefile index 1949fe5..18d56d8 100644 --- a/arch/arm64/kvm/Makefile +++ b/arch/arm64/kvm/Makefile @@ -27,3 +27,4 @@ kvm-$(CONFIG_KVM_ARM_HOST) += $(KVM)/arm/vgic-v3.o kvm-$(CONFIG_KVM_ARM_HOST) += $(KVM)/arm/vgic-v3-emul.o kvm-$(CONFIG_KVM_ARM_HOST) += vgic-v3-switch.o kvm-$(CONFIG_KVM_ARM_HOST) += $(KVM)/arm/arch_timer.o +kvm-$(CONFIG_KVM_ARM_PMU) += $(KVM)/arm/pmu.o diff --git a/include/kvm/arm_pmu.h b/include/kvm/arm_pmu.h index ddcb5b2..14bedb0 100644 --- a/include/kvm/arm_pmu.h +++ b/include/kvm/arm_pmu.h @@ -34,9 +34,20 @@ struct kvm_pmu { int irq_num; struct kvm_pmc pmc[ARMV8_MAX_COUNTERS]; }; + +u64 kvm_pmu_get_counter_value(struct kvm_vcpu *vcpu, u64 select_idx); +void kvm_pmu_set_counter_event_type(struct kvm_vcpu *vcpu, u64 data, + u64 select_idx); #else struct kvm_pmu { }; + +u64 kvm_pmu_get_counter_value(struct kvm_vcpu *vcpu, u64 select_idx) +{ + return 0; +} +void kvm_pmu_set_counter_event_type(struct kvm_vcpu *vcpu, u64 data, + u64 select_idx) {} #endif #endif diff --git a/virt/kvm/arm/pmu.c b/virt/kvm/arm/pmu.c new file mode 100644 index 0000000..b107fb8 --- /dev/null +++ b/virt/kvm/arm/pmu.c @@ -0,0 +1,122 @@ +/* + * Copyright (C) 2015 Linaro Ltd. + * Author: Shannon Zhao <shannon.zhao@linaro.org> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see <http://www.gnu.org/licenses/>. + */ + +#include <linux/cpu.h> +#include <linux/kvm.h> +#include <linux/kvm_host.h> +#include <linux/perf_event.h> +#include <asm/kvm_emulate.h> +#include <kvm/arm_pmu.h> + +/** + * kvm_pmu_get_counter_value - get PMU counter value + * @vcpu: The vcpu pointer + * @select_idx: The counter index + */ +u64 kvm_pmu_get_counter_value(struct kvm_vcpu *vcpu, u64 select_idx) +{ + u64 counter, reg, enabled, running; + struct kvm_pmu *pmu = &vcpu->arch.pmu; + struct kvm_pmc *pmc = &pmu->pmc[select_idx]; + + reg = (select_idx == ARMV8_CYCLE_IDX) + ? PMCCNTR_EL0 : PMEVCNTR0_EL0 + select_idx; + counter = vcpu_sys_reg(vcpu, reg); + + /* The real counter value is equal to the value of counter register plus + * the value perf event counts. + */ + if (pmc->perf_event) + counter += perf_event_read_value(pmc->perf_event, &enabled, + &running); + + return counter & pmc->bitmask; +} + +static inline bool kvm_pmu_counter_is_enabled(struct kvm_vcpu *vcpu, + u64 select_idx) +{ + return (vcpu_sys_reg(vcpu, PMCR_EL0) & ARMV8_PMCR_E) && + (vcpu_sys_reg(vcpu, PMCNTENSET_EL0) & BIT(select_idx)); +} + +/** + * kvm_pmu_stop_counter - stop PMU counter + * @pmc: The PMU counter pointer + * + * If this counter has been configured to monitor some event, release it here. + */ +static void kvm_pmu_stop_counter(struct kvm_vcpu *vcpu, struct kvm_pmc *pmc) +{ + u64 counter, reg; + + if (pmc->perf_event) { + counter = kvm_pmu_get_counter_value(vcpu, pmc->idx); + reg = (pmc->idx == ARMV8_CYCLE_IDX) + ? PMCCNTR_EL0 : PMEVCNTR0_EL0 + pmc->idx; + vcpu_sys_reg(vcpu, reg) = counter; + perf_event_release_kernel(pmc->perf_event); + pmc->perf_event = NULL; + } +} + +/** + * kvm_pmu_set_counter_event_type - set selected counter to monitor some event + * @vcpu: The vcpu pointer + * @data: The data guest writes to PMXEVTYPER_EL0 + * @select_idx: The number of selected counter + * + * When OS accesses PMXEVTYPER_EL0, that means it wants to set a PMC to count an + * event with given hardware event number. Here we call perf_event API to + * emulate this action and create a kernel perf event for it. + */ +void kvm_pmu_set_counter_event_type(struct kvm_vcpu *vcpu, u64 data, + u64 select_idx) +{ + struct kvm_pmu *pmu = &vcpu->arch.pmu; + struct kvm_pmc *pmc = &pmu->pmc[select_idx]; + struct perf_event *event; + struct perf_event_attr attr; + u64 eventsel, counter; + + kvm_pmu_stop_counter(vcpu, pmc); + eventsel = data & ARMV8_EVTYPE_EVENT; + + memset(&attr, 0, sizeof(struct perf_event_attr)); + attr.type = PERF_TYPE_RAW; + attr.size = sizeof(attr); + attr.pinned = 1; + attr.disabled = kvm_pmu_counter_is_enabled(vcpu, select_idx); + attr.exclude_user = data & ARMV8_EXCLUDE_EL0 ? 1 : 0; + attr.exclude_kernel = data & ARMV8_EXCLUDE_EL1 ? 1 : 0; + attr.exclude_hv = 1; /* Don't count EL2 events */ + attr.exclude_host = 1; /* Don't count host events */ + attr.config = eventsel; + + counter = kvm_pmu_get_counter_value(vcpu, select_idx); + /* The initial sample period (overflow count) of an event. */ + attr.sample_period = (-counter) & pmc->bitmask; + + event = perf_event_create_kernel_counter(&attr, -1, current, NULL, pmc); + if (IS_ERR(event)) { + printk_once("kvm: pmu event creation failed %ld\n", + PTR_ERR(event)); + return; + } + + pmc->perf_event = event; +}